Evaluating ECMWF global model cloud and precipitation fields with observed radar reflectivity: How do we ensure a fair comparison?

Poster PDF

Description

Remote sensing data from ground-based and space-borne radar and lidar provide a wealth of information on cloud and precipitation that can be used to evaluate atmospheric models. One common approach to take account of the different quantities and spatial and temporal scales between model and observations is to use a forward operator to derive the observed quantity using information from the model. However, it is vital to understand the limitations and uncertainties of the comparison in order to separate real deficiencies in the model from artefacts of the forward model. Here we use the ECMWF global Numerical Weather Prediction model and radar reflectivity data from the CloudSat and ARM radars to assess the sensitivity of uncertainties in the radar reflectivity forward model. We highlight issues relating to microphysical assumptions, representation of hydrometeors in discrete categories, and sub-grid scale heterogeneity that need to be taken into account when evaluating model cloud and precipitation fields with radar reflectivity observations.